funasr/bin/eend_ola_inference.py
@@ -209,7 +209,7 @@ if data_path_and_name_and_type is None and raw_inputs is not None: if isinstance(raw_inputs, torch.Tensor): raw_inputs = raw_inputs.numpy() data_path_and_name_and_type = [raw_inputs[0], "speech", "bytes"] data_path_and_name_and_type = [raw_inputs[0], "speech", "sound"] loader = EENDOLADiarTask.build_streaming_iterator( data_path_and_name_and_type, dtype=dtype,
